How to build a very simple delayed power oft circuit?

**Arathranar** · 01-22-2005, 05:24 PM

All the threads to do with shut down controllers have been overkill for my needs.

I just need a circuit to cut a power line about 60 seconds (or more) after it sees the switched accessory line go to ground (it's also at ground during crank), but reconnect the line immediately at ignition on (acc line goes high). I have a tank circuit to supply power to the PC in those 60 seconds but want to avoid the tank battery and eventually the car battery being drained by the 70mA of current draw by the PC in in hibernate or off.

**Curiosity** · 01-22-2005, 06:46 PM

If you're confident that your PC will shut down, then all you need is a relay and 2 diodes. We've talked about this a bunch. Either ACC or the PCs 12V powers the relay, so as soon as the PC starts up with ACC on, it holds the relay until the PC completely shuts down and then kills it's own power.

**Arathranar** · 01-22-2005, 08:03 PM

I'm intending to use the fake UPS trick to shut down the PC after a 30 seconds delay. I *believe* that shut down due to a UPS will always force the shutdown. And I was slightly wrong above - my tank circuit is hooked up the ACC line so the PC can, at worst, empty the tank battery only.

So acc->diode->relay & PC->diode->same relay. Sounds good - that's about as simple as I could hope for

. Cunning circuit - I didn't see that mentioned anywhere in my search of the forum. Thanks.

I just realized if I want the PC to also automatically turn on, I also need a circuit to toggle the power button line into the m/b (I don't have an option in the bios to enable auto turn-on). I can drive that from the same circuit at the realy side of the diodes since that'll only go +ve once despite the acc line going low during crank. Is there anything simple that will momentarily short the power button line on a 0->12V transition only?

**Curiosity** · 01-23-2005, 03:03 AM

That's exactly what I did. A constant to momentary on the part that powers the relay. It's on www.the12volt.com and a few places here as well. I used a 12V reed relay, 470uF el cap, and 15K resistor all from radio shack. Look at the part on the right of this. PwrButton goes to the relay. I should have put that there. Also, the relay needs a clamping diode. I found out what that can do if it's not there. When the PC is off and you turn the key off, it closes the relay for about 1 second and starts the PC up. Not good.

**Arathranar** · 01-23-2005, 03:31 AM

Ah yes. That thread! Is that circuit still working well for you?

**Arathranar** · 01-23-2005, 03:36 AM

By clamping diode do you mean this?

**Arathranar** · 01-23-2005, 03:39 AM

Where does the other power button pin go?

**Arathranar** · 01-23-2005, 03:52 AM

I'm confused. I can see how the constant to momentary circuit on the12Volt works. But how does your circuit work without the relay?

**Curiosity** · 01-23-2005, 03:52 AM

I'm still using that and it works fine, but it'll get replaced soon enough. I'm thinking it was a total waste of time when a simple mosfet would keep the voltage from dropping much. Shoulda listened more!

That's the right part of the circuit you've got though. The relay that's not there would be connected to PwrButton and would have a clamping diode on it. You don't need one on the other relay since it has nowhere to feedback to. Then the contacts connect to the PCs power button.

**Arathranar** · 01-23-2005, 03:56 AM

Originally Posted by Curiosity

That's the right part of the circuit you've got though. The relay that's not there would be connected to PwrButton and would have a clamping diode on it. You don't need one on the other relay since it has nowhere to feedback to. Then the contacts connect to the PCs power button.

That makes a lot more sense! The other side of that relay's coil is just to ground?